Instead of using traditional approaches in the audio domain to defend against attacks, this team will explore capturing voice commands in another domain (e.g., vibration domain) and then perform cross-domain authentication. For example, sensors embedded in wearable devices (e.g., smart watches or activity trackers) will capture users’ speech signals and compare them with audio signals to authenticate the user.
Project 20-11-01: Securing Voice Control Technology in Manufacturing with Cross-Domain Authentication
Published on October 3, 2023
PROJECT AWARDED: November 1, 2021; PROJECT LEAD: Rutgers University; PROJECT PARTICIPANTS: Texas A&M Engineering Experiment and Station (TEES)
Rutgers is partnering with Texas A&M Engineering Experiment Station (TEES) to develop new methods to secure voice control technology and help manufacturers defend themselves against voice-based cyber attacks.
Problem
Proposed Solution
The team will develop and validate a digital tool that collects data on processes and quality. That data will then provide direct feedback to the operator and process planner, assisting them as they make decisions about quality control. The team plans to use sensor technology with machine learning algorithms that will incrementally train a predictive model. That model can then be consistently updated to predict quality measures based on process input parameters.
Impact
This project will advance the security of voice control technology in manufacturing by using cross-domain sensing capabilities to enhance user authentication. New software will be available to deploy, and new signal processing and machine learning techniques will be developed to study various sensing signals.